Zhang 2015 how do hosts react t

Interestingly, unlike certain Wolbachia transinfections in novel hosts, the Wolbachia–host association in the present study showed no clear evidence of host immune priming by Wolbachia,

How do hosts react to endosymbionts? A new insight into the molecular mechanisms underlying the

Wolbachia–host association

Y.-K Zhang, X.-L Ding, X Rong and X.-Y Hong

Department of Entomology, Nanjing Agricultural

University, Nanjing, China

Abstract

Wolbachia is an intracellular bacterium that has

aroused intense interest because of its ability to alter

the biology of its host in diverse ways In the

two-spotted spider mite, Tetranychus urticae, Wolbachia

can induce complex cytoplasmic incompatibility (CI)

phenotypes and fitness changes, although little is

known about the mechanisms In the present study,

we selected a strain of T urticae, in which Wolbachia

infection was associated with strong CI and enhanced

female fecundity, to investigate changes in the

transcriptome of T urticae in Wolbachia-infected vs.

uninfected lines The responses were found to be

sex-specific, with the transcription of 251 genes being

affected in females and 171 genes being affected in

males Some of the more profoundly affected genes in

both sexes were lipocalin genes and genes involved

in oxidation reduction, digestion and detoxification.

Several of the differentially expressed genes have

potential roles in reproduction Interestingly, unlike

certain Wolbachia transinfections in novel hosts, the

Wolbachia–host association in the present study

showed no clear evidence of host immune priming by

Wolbachia, although a few potential immune genes

were affected.

Keywords: two-spotted spider mite, Wolbachia,

transcriptome sequencing, gene expression.

Introduction

Bacterial intracellular symbiosis is widespread in inver-tebrates Many intracellular bacteria are known to influ-ence host biological processes, including developmental

programmes, reproduction and immunity (Duron et al., 2008; Vallet-Gely et al., 2008; Tsuchida et al., 2010; Ivanov & Littman, 2011) The bacterium Wolbachia is

perhaps the most abundant vertically transmitted microbe worldwide, infecting an estimated 40% of terrestrial arthropods (Zug & Hammerstein, 2012) Interestingly,

Wolbachia can induce a range of reproductive

manipula-tions in arthropods that facilitate vertical transmission

(Werren et al., 2008) In addition to their effects on host reproduction, Wolbachia have mutualistic relationships with nematodes (Taylor et al., 2013) and increase the

resistance of mosquitoes and flies to various pathogens, such as dengue fever virus, Chikungunya virus and yellow

fever virus (Hedges et al., 2008; van den Hurk et al.,

2012).

Many studies have examined the phenotypic effects of

Wolbachia infection on host physiology and immunity.

Recent studies have begun to clarify the molecular

mechanisms underlying these effects In Drosophila melanogaster, the finding that most of the genes in larval

testes putatively associated with reproduction (especially

spermatogenesis) were downregulated by Wolbachia

may help elucidate the underlying mechanisms of

Wolbachia-induced cytoplasmic incompatibility (CI)

(Zheng et al., 2011) In the wasp Asobara tabida, Wolbachia is required for oogenesis (Dedeine et al.,

2005), possibly because of its interference with the

expression of ferritin (Kremer et al., 2009, 2012) In the fruit flies D melanogaster and Drosophila simulans, Wolbachia confers resistance against RNA viral infection (Hedges et al., 2008; Teixeira et al., 2008; Osborne et al., 2009) In the mosquito Aedes aegypti, Wolbachia confers

resistance against various pathogens notably by priming

the innate immune system (Moreira et al., 2009; Bian

et al., 2010) and affects the expression of microRNAs

First published online 15 September 2014

Correspondence: Prof X.-Y Hong, Department of Entomology, Nanjing

Agricultural University, Nanjing, Jiangsu 210095, China Tel & fax: 0086 25

84395339; e-mail: xyhong@njau.edu.cn

Biology

Insect Molecular Biology (2015) 24(1), 1–12 doi: 10.1111/imb.12128

(Hussain et al., 2011; Zhang et al., 2013) Taken together,

these findings clearly indicate that Wolbachia can

influ-ence host gene transcription in ways that increase its

own survival Because the effects of Wolbachia on host

biology are strain- and host-specific (Serbus et al., 2008),

studies of additional invertebrate systems are needed to

unravel the conserved and diverged mechanisms in

host–Wolbachia interactions.

The spider mite Tetranychus urticae is a cosmopolitan

agricultural pest with an extensive host plant range and

strong pesticide resistance (Bolland et al., 1998) Its

genome, at 90 megabases, is the smallest sequenced

arthropod genome (Grbic´ et al., 2011) It is also the

first complete genome among the Chelicerates, which

diverged from other arthropod lineages more than 450

Mya (Dunlop, 2010) Wolbachia is widely distributed in

T urticae, in which it can induce anything from no CI to

complete CI (Breeuwer, 1997; Perrot-Minnot et al., 2002;

Vala et al., 2002; Gotoh et al., 2007) Wolbachia can also

affect the fitness of T urticae For example, in mites

col-lected from cucumber plants, the offspring of

Wolbachia-infected females had higher survival rates (Vala et al.,

2003), and in some Chinese populations, Wolbachia

infection increased female fecundity (Xie et al., 2011;

Zhao et al., 2013) Although gene expressions in

T urticae are affected by host plant transfer and

diapause (Dermauw et al., 2012; Bryon et al., 2013), little

is known about how they are affected by Wolbachia

infection.

In the present study, we explored the effects of

Wolbachia on the T urticae transcriptome by comparing

gene expression profiles in infected and uninfected adult

mites of a strain whose CI and fecundity are strongly

affected by Wolbachia (Zhao et al., 2013) Wolbachia was

found to affect the expressions of many genes, including

genes involving oxidation reduction, digestion

detoxifica-tion and reproducdetoxifica-tion These results provide new insights

for an understanding of the complex interactions between

arthropods and Wolbachia.

Results

Sequence data processing

We sequenced four transcriptome libraries (Turt_FI, Turt_FU, Turt_MI and Turt_MU), each with between 55 and 66 million reads About 79% of the reads in each group were uniquely mapped to the reference genome (Table 1) A total of 18 813 genes, including 18 204 known

T urticae genes and 519 novel genes, were detected in

the four libraries To better evaluate the transcriptional status between libraries, the gene expression levels were divided into five grades according to their reads per kilobase of exon model per million mapped reads (RPKM) values Results indicated that most genes were expressed

at low levels in all libraries, samples from the same sex had a similar gene expression pattern at each RPKM interval (Fig 1).

Transcriptional responses to Wolbachia infection

A global view of gene expressions in the four libraries is presented in the hierarchical clustering heat map in Fig 2A As shown, gene expressions are sex-specific Comparison of Turt_MI and Turt_MU and comparison of Turt_FI and Turt_FU show that each library has unique transcriptional changes, suggesting that the expressions

of many genes are affected by Wolbachia infection, although we cannot exclude non-Wolbachia related

genetic differences between lines that may have accu-mulated during the antibiotic curing and selection process, or antibiotic-mediated changes in gut bacterial composition.

In females, 148 genes were observed to be upregulated

and 103 downregulated in the Wolbachia-infected line compared with the Wolbachia-uninfected line, while in

males, 96 genes were upregulated and 75 downregulated

in the Wolbachia-infected line In addition, 82 genes were

differentially expressed in both females and males com-pared with uninfected ones (Fig 2B, Table S2A, S2B).

Table 1 Summary of sequencing results of Tetranychus urticae transcriptome

Turt_FI, infected females; Turt_FU, uninfected females; Turt_MI, infected males; Turt_MU, uninfected males

Figure 1 Distribution of Tetranychus urticae unigenes The y-axis represents gene number; the x-axis represents reads per kilobase of exon model per

million mapped reads (RPKM) range of genes Turt_FI, infected females; Turt_FU, uninfected females; Turt_MI, infected males; Turt_MU, uninfected males

Figure 2 Tetranychus urticae transcriptional responses to Wolbachia infection A Hierarchical clustering heat map of the gene abundance in the four

samples Colours from blue to red represent the gene expression abundance from poor to rich B Venn diagram showing significant gene expression

change in response to Wolbachia infection in T urticae females and males Turt_FI, infected females; Turt_FU, uninfected females; Turt_MI, infected

males; Turt_MU, uninfected males

In females, the strongest changes in the

Wolbachia-infected line were in the biological process gene ontology

(GO) categories of the oxidation-reduction process, the

chlorophyll metabolic process, virus–host interaction,

interaction with the host and pigment metabolic process

(Fig 3A) and in the following molecular function GO

cat-egories: carboxylic ester hydrolase activity; 4 iron cluster

binding; 4 sulphur cluster binding; and oxidoreductase

activity In males, the strongest changes were in the

fol-lowing GO categories of the oxidation-reduction

pro-cess: 4 iron cluster binding; 4 sulfur cluster binding; and

oxidoreductase activity (Fig 3B) Interestingly, no GO

categories related to host immune priming were found to

be affected in the Wolbachia-infected line in adults of

either sex, although a few potential immune genes were

affected Other GO categories of differentially expressed

genes in females and males are listed in Table S3A and

S3B, respectively.

Kyoto Encyclopaedia of Genes and Genomes

pathway analysis

In females, genes that were differentially expressed in

the Wolbachia-infected line were involved in 44 Kyoto

Encyclopaedia of Genes and Genomes (KEGG)

path-ways, mainly involving lysosome process, phenylalanine

metabolism and valine, leucine and isoleucine degradation

(Table S4A) In males the differentially expressed genes

were involved in 45 KEGG pathways, mainly involving

fatty acid metabolism, β-alanine metabolism, propanoate

metabolism and lysosome process (Table S4B).

Differentially expressed genes of interest

Tetranychus urticae and other spider mites are known to

express several genes for detoxifying enzymes, including

cytochrome P450 monooxygenases (CYPs),

glutathione-S-transferases and ABC transporters Wolbachia affected

the expression of seven CYPs in females and five

CYPs in males (Table 2) Most were downregulated Two

glutathione-S-transferase genes were downregulated and

five ABC transporter genes were upregulated.

Lipocalins, which are small proteins capable of binding

to hydrophobic molecules, were mostly downregulated

(eight of ten in females, five of six in males; Table 2) All

the differentially expressed ribosomal proteins were

downregulated in females The expression of many genes

of unknown function changed dramatically in females

(Table S2A) and in males (Table S2B) The products of

many of these genes are predicted to be secreted or

conserved hypothetical proteins Among genes potentially

associated with oogenesis or embryogenesis, the gene

for LKAP32 limkain-b1 (tetur11g00940), which interferes

with meiosis, and two genes encoding vitellogenin

(tetur20g01230, tetur39g00740), were upregulated Two

genes, tetur13g00510 and tetur03g01690, which are

involved in chromatin assembly, were upregulated and downregulated, respectively Remarkably, seven genes encoding cuticular protein were all upregulated (Table 2), possibly reflecting disruption of embryogenesis.

The expressions of 64 putative immunity-related genes involved in the Toll, Imd/Jnk and JAK-STAT pathways

were not significantly affected in the Wolbachia-infected

line The GO enrichment analysis confirmed these results.

Three cystatin genes (tetur06g06620, tetur09g03670 and tetur09g04770) were differentially expressed in females.

The former was downregulated whereas the latter two

were upregulated Another cystatin gene (tetur06g01060)

was downregulated in males Two of the genes in the

autophagy pathway (tetur07g07470 and tetur24700010)

encode ATG4 autophagy related 4 homologue A The former was upregulated in box sexes and the latter was downregulated (Table 2).

Quantitative real-time PCR validation

Eighteen differentially expressed genes (ten in females, eight in males) were randomly selected to validate the expression profiles obtained with the RNA-Seq analysis All of them yielded PCR products, whose sequences matched the RNA-Seq generated sequences perfectly The changes in expression of all but two of the genes

(tetur11g00940 and tetur04g02680) were in good

agree-ment with the RNA-Seq results Among these genes in

females, tetur13g00510 showed the largest upregula-tion and tetur97g00020 manifested the largest

down-regulation, and tetur26g01450 showed the largest downregulation in males, which was consistent with the RNA-seq results (Fig 4) These results demonstrate the reliability of the RNA-Seq results.

Discussion

Wolbachia widely infects arthropods and can have

important consequences for the fitness of their hosts.

Wolbachia can interact with their invertebrate hosts at both the molecular and cellular levels (Xi et al., 2008; Kremer et al., 2009, 2012; Yamada et al., 2011) These

findings, together with the sequencing of genomes of

Wolbachia strains that induce various phenotypic effects (Wu et al., 2004; Klasson et al., 2008; Salzberg et al., 2009; Darby et al., 2012) have greatly clarified the evolv-ing relationship between Wolbachia and their hosts Although Wolbachia has various phenotypic effects on

T urticae, the mechanisms are not well understood In the present study, we selected a strain of T urticae in which Wolbachia induces strong CI and increases female fecun-dity (Zhao et al., 2013) to investigate its responses to Wolbachia infection Wolbachia increases the fecundity of

Figure

D mauritiana and the mitotic activity of germline stem

cells, as well as decreases programmed cell death in the

germarium (Fast et al., 2011) The strength of CI induced

by Wolbachia infection dramatically decreased with both

male age (Reynolds & Hoffmann, 2002) and larval stage

development (Yamada et al., 2007) As a result, we

hypothesized that Wolbachia would strongly affect

female fecundity and early spermatogenesis in T urticae.

In the present study, newly emerged adult females

and 1-day-old adult virgin males were collected for transcriptome analysis to test this hypothesis.

Most of the transcriptome reads that we obtained could

be mapped to the reference genome The transcripts cor-responded to both known genes and some novel genes Our findings suggest that the northeast China strain used

in the present study genetically differs from the Canadian

strain used for the genome project (Grbic´ et al., 2011),

which is not surprising because mite strains are known

Table 2 Candidate genes differentially expressed in response to Wolbachia infection in Tetranychus urticae

log2(FC) P value log2(FC) P value

Digestion or Detoxification tetur03g00830 Cytochrome P450-CYP392A12 −1.06 9.03E-10 −1.71 1.33E-35

tetur26g01450 Glutathione S-transferase class delta −6.68 1.93E-08 −7.63 1.13E-28 tetur26g01460 Glutathione S-transferase class delta −5.9 9.12E-09 −5.9 1.74E-08

tetur31g00710 Apolipoprotein D probably a pseudogene −2.34 6.18E-10 −2.83 1.64E-33

tetur247g00010 ATG 4 autophagy related 4 homolog A −2.27 4.09E-37 −2.34 6.24E-05

Genes are ranked by biological process and/or molecular function Gene IDs and descriptions were compiled from the T urticae genome project FC, fold

change

to be genetically diverse (Grbic´ et al., 2011) In addition,

novel transcripts can be detected with increasing

sequencing depth and coverage (Sims et al., 2014).

The main objective of the present study was to

detect differentially expressed processes in response to

Wolbachia infection The high depth sequencing made it

possible to analyse the libraries at the gene level The

expression patterns of males and females differed in the

Wolbachia-infected vs the uninfected lines, suggesting

that there are specific differences between the sexes The

number of genes regulated by Wolbachia infection is

prob-ably related to the cellular tropism and virulence of the

strain of Wolbachia as well as its density in the host (Walker et al., 2011; Rancès et al., 2012).

A notable finding of the GO analysis was the enrichment

of gene sets related to oxidoreductase activity in both sexes Oxidoreductase is involved in energy metabolism and redox homeostasis One consequence of disrupted redox homeostasis is DNA damage, which includes single- and double-stranded breaks, base and

deoxyri-bose modifications, and DNA cross-linking (Valko et al., 2006; Brennan et al., 2012) Wolbachia has been shown

to disturb the cellular physiology of its insect host

espe-cially via the generation of oxidative stress (Brennan et al.,

Figure 4 Validation of RNA-sequencing data by real-time quantitative-PCR analysis in females (A) and males (B) Fold differences in the expression of

selected genes in response to Wolbachia infection The fold differences were calculated using the 2−ΔΔCtmethod Data are presented as mean±SD

values of triplicate reactions for each gene transcript

2008; Pan et al., 2012) Similarly, Wolbachia induces an

increase in antioxidant expression in mosquito cells,

which could be an adaptation to symbiosis Moreover, in

A tabida, Wolbachia interferes with iron metabolism,

which limits oxidative stress and cell death, thus

promot-ing its survival within host cells (Kremer et al., 2010) Our

finding that multiple genes are involved in oxidation

reduc-tion raises the possibility that Wolbachia regulates redox

reactions to reduce reactive oxygen species levels and

thus maintain the Wolbachia–host symbiotic relationship;

however, in T urticae, oxidoreductase activity was found

to be associated with host plant transfer and diapause

within T urticae (Grbic´ et al., 2011; Bryon et al., 2013),

which raises the possibility that spider mites respond to

stress by regulating oxidoreductase activity.

T urticae is among the most polyphagous herbivores

and harbours a large number of detoxification genes In

the present study, a set of these specialized genes was

profoundly affected by Wolbachia infection, suggesting

that Wolbachia affects spider mite feeding and

detoxi-fication Many lipocalins genes were downregulated in

response to Wolbachia infection Lipocalins are small

extracellular proteins that typically bind hydrophobic

molecules (Chudzinski-Tavassi et al., 2010) In spider

mites, they may bind pesticides/allelochemicals, resulting

in sequestration of these toxic, generally hydrophobic

compounds The fact that Wolbachia is widely distributed

in natural populations of T urticae (Breeuwer, 1997;

Perrot-Minnot et al., 2002; Vala et al., 2002; Gotoh et al.,

2007) raises the possibility that it has a role in T urticae

resistance to diverse plant chemicals and pesticides.

Further studies are needed to check this possibility Many

of the genes that were differentially transcribed in the

Wolbachia-infected vs the uninfected lines encode

teins that are secreted Although the roles of these

pro-teins are unclear, the finding that Wolbachia are mainly

located in the gnathosoma in both sexes (Zhao et al.,

2013) may indicate that the affected genes are involved in

the digestion and detoxification of food Among the genes

affected by Wolbachia, many have no known function For

example, 71 of these genes were unique to female

T urticae and 63 were unique to male T urticae.

Although Wolbachia induces strong CI and enhances

female fecundity in T urticae, we didn’t identify any

affected genes that were related to oogenesis or

sper-matogenesis; however, some of the genes may be related

to female reproduction For instance, two genes encoding

vitellogenin, were upregulated in infected females.

Vitellogenins are important for growth and differentiation

of oocytes and transporting metallic ions, lipids and

vita-mins into the oocytes (Raikhel & Dhadialla, 1981); hence,

these genes might have a role in enhancing female

fecun-dity Three other genes may have roles in meiosis, as two

of them are involved in chromatin assembly, and one is

involved in meiosis arrest Another set of upregulated

genes encoded cuticle proteins In the nematode Brugia malayi, removal of Wolbachia downregulated transcripts

involved in cuticle biosynthesis, possibly reflecting a dis-ruption of the normal embryogenic programme (Ghedin

et al., 2009).

Several genes that may participate in CI have been

identified in D melanogaster (Xi et al., 2008; Zheng et al., 2011) and D simulans (Clark et al., 2006; Landmann

et al., 2009) In D melanogaster, male development time

was found to be inversely correlated with the strength of

CI (Yamada et al., 2007) In the present study, a gene (tetur10g04620) that encodes juvenile hormone-binding

protein was found to be downregulated in infected males The presumably higher expression of this protein in uninfected males would allow the testes to develop com-pletely and produce fully mature sperm that would not be

able to induce CI In support of this idea, Zheng et al (2011) found that, in D melanogaster, Wolbachia infection

resulted in a ∼10-fold increase in the transcription of the gene for juvenile hormone-induced protein (JhI-26) in the testes of late-stage larvae.

The host’s immune system is pivotal to maintaining a balanced relationship with an endosymbiont There is growing evidence that the presence of a symbiont can dramatically affect host immunity (reviewed by Gross

et al., 2009); however, we found that the expression levels

of almost all of the putative immunity-related genes were

stable during Wolbachia infection The expression of only

a few genes involved in humoral immunity and the

autophagy pathway was altered in the Wolbachia-infected

vs the uninfected lines, which is strikingly different from

what was found in other host–Wolbachia associations (Chevalier et al., 2012; Kremer et al., 2012; Rancès et al., 2012) Although the T urticae genome has genes that are

involved in the Toll and Imd pathways, it lacks other com-ponents that are essential for effective immune signalling The genome also lacks extracellular serine proteases, putative phenoloxidase and most of the antimicrobial

peptide effector gene orthologues (Grbic´ et al., 2011) The repertoire of immunity genes found in T urticae is

consist-ent with a pattern emerging from comparative studies in invertebrate immunity, suggesting that invertebrates use diverse solutions to build an immune response, probably

driven by specific life histories (Loker et al., 2004) It is also possible that Wolbachia adopts a strategy for

escap-ing from the host immune system in the evolutionary

rela-tionship with T urticae Wolbachia, being intracellular, are

surrounded by cell membranes that may protect them from the host immune system To date, immune priming by

Wolbachia has mainly been observed in experiments

on heterologous host systems The immune gene upregulation has been observed in novel

laboratory-generated transinfections of naturally Wolbachia-free

species, not in long-established natural associations such

as the one studied here Identifying differences in host

immune response induction among different Wolbachia

strains will help to clarify the interactions between

Wolbachia and their hosts (Bourtzis et al., 2000; Chevalier

et al., 2012; Kremer et al., 2012; Rancès et al., 2012).

In summary, Wolbachia infection affects numerous

biological processes in T urticae, including oxidation

reduction processes, digestion and detoxification, and

processes involving lipocalins Unexpectedly, we found no

evidence for strong effects of Wolbachia infection on

spider mite reproduction and immunity As a study of

the molecular mechanisms underlying the Chelicerata–

Wolbachia association, this work provided new insights for

understanding the complex interactions between

arthro-pods and Wolbachia.

Experimental procedures

Mite rearing and sample collection

Mites used in the present study were originally collected from

Hohhot, Inner Mongolia, northeast China in July 2010 and reared

on leaves of the common bean (Phaseolus vulgaris L.) placed on

a water-saturated sponge mat in Petri dishes at 25± 1°C, 60%

relative humidity and under 16 h light: 8 h dark conditions To

establish 100% infected and 100% uninfected Wolbachia lines

with identical genetic backgrounds, one female from the

teleiochrysalis stage was allowed to lay eggs without being

crossed with males The eggs were reared until adulthood

(males) After the males had reached sexual maturity, they were

backcrossed with the mother After the cross, the female adults

were transferred to new leaf discs and were allowed to lay eggs

for 3–5 days Females were each checked for Wolbachia

infec-tion by PCR amplificainfec-tion The eggs were reared separately on

new leaf discs depending on the infection status of the mother

The above process was continued for four generations until all

members of the population were confirmed to be infected with

Wolbachia (a 100% singly Wolbachia-infected population was

obtained) The uninfected lines were established by treating lines

singly infected with Wolbachia with tetracycline Small leaf discs

(∼3 cm2) from the common bean were placed on a cotton bed

soaked in tetracycline solution (0.1%, w/v) in Petri dishes (9 cm in

diameter), and kept for 24 h before they were used for rearing the

newly hatched larvae Distilled water was added daily to keep the

cotton beds wet The cotton and the leaf discs were replaced

every 4 days Three generations later, mites were checked using

PCR to confirm that the lines were free of Wolbachia These lines

were maintained in a mass-rearing environment without

antibiot-ics for approximately four generations (2 months) before use, to

avoid the potential side effects of antibiotic treatment Through

PCR assays, neither line was found to be infected with Cardinium

(Primers: CLO-f1: 5

′-GGAACCTTACCTGGGCTAGAATGTATT-3′, CLO-r1: 5′-GCCACTGTCTTCAAGCTCTACCAAC-3′) or

Rickettsia (Primers: R1: 5′-GCTCTTGCAACTTCTATGTT-3′, R2:

5′-CATTGTTCGTCAGGTTGGCG-3′) (Duron et al., 2008), which

can manipulate host reproduction Within the two lines,

1–3-day-old adult females (Turt_FI and Turt_FU, ‘I’ indicates infection and

‘U’ indicates uninfection) and 1-day-old adult virgin males

(Turt_MI and Turt_MU) were respectively collected The samples were stored in liquid nitrogen until required for RNA isolation

Library construction and RNA sequencing

Total RNA was extracted using the Trizol protocol (Invitrogen, Carlsbad, CA, USA) and RNA quality was determined by an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) according to the manufacturer’s recommendations Poly (A) mRNA was isolated with oligo-dT beads and then treated with the fragmentation buffer The cleaved RNA fragments were then tran-scribed into first-strand cDNA using reverse transcriptase and random hexamer primers This was followed by second-strand cDNA synthesis using DNA polymerase I and RNaseH The double-stranded cDNA was further subjected to end repair using T4 DNA polymerase, Klenow fragment DNA polymerase I, and T4 polynucleotide kinase, followed by a single A base addition using Klenow 3′ to 5′ exo-polymerase It was then ligated with an adapter or index adapter using T4 quick DNA ligase Adaptor-ligated fragments were selected according to the size and the desired range of cDNA fragments was excised from the gel PCR was performed to selectively enrich and amplify the fragments Finally, after validating the fragment quality on an Agilent 2100 Bioanalyzer (Agilent Technologies) and ABI Step One plus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA), the cDNA library was sequenced on a flow cell using Illumina HiSeq2000 (San Diego, CA, USA)

Sequencing data quality control

After Illumina sequencing, a sequence-filtering process was used

to select clean reads First, Illumina’s Failed-Chastity filter soft-ware was used to remove raw reads that fell into the relation

‘failed-chastity≤1’, with a chastity threshold of 0.6 on the first 25 cycles Second, all raw reads showing signs of adaptor contami-nation or ambiguous trace peaks (denoted with an ‘N’ in the sequence trace) were removed Finally, raw reads showing

>10% of bases with a Phredscaled probability (Q) <20 were discarded All the downstream analyses were based on the clean data with high quality

Reads mapping to the reference genome

Reference genome and gene model annotation files were

down-loaded from the T urticae genome website (http://bioinformatics

.psb.ugent.be/webtools/bogas/overview/Tetur) directly An index

of the reference genome was built using BOWTIE v2.0.6 (Langmead & Salzberg, 2012) and paired-end clean reads were aligned to the reference genome using TOPHATv2 0 7 (Trapnell

et al., 2009) We selected TOPHATas the mapping tool because it can generate a database of splice junctions based on the gene model annotation file, which other non-splice mapping tools cannot do

Quantification of gene expression level

HTSEQv0.5.3 (Anders & Huber, 2011) was used to count the reads mapped to each gene In this study we used ‘reads per kilobase of exon model per million mapped reads’ (RPKM), which considers the effect of sequencing depth and gene length for the reads count at the same time, and is currently the most commonly used method for estimating gene expression levels (Mortazavi

et al., 2008) For each gene, the RPKM value was calculated

based on the length of the gene and the number of reads mapped

to the gene

Differential expression analysis

For each sequenced library, the read counts were adjusted by the

EDGER program package (Robinson et al., 2010) through one

scaling normalized factor Differential expression analysis of two

samples (Turt_FI and Turt_FU, Turt_MI and Turt_MU) was then

analysed using the DEGSeq R package 1.12.0 (Anders & Huber,

2012) The P values were adjusted using the Benjamini &

Hochberg (1995) method A corrected P value of 0.005 and a log2

(fold change) of 1 were set as the threshold for significant

differ-ential expression

Gene ontology and Kyoto Encyclopedia of Genes and

Genomes enrichment analysis of differentially expressed genes

Gene ontology enrichment analysis of differentially expressed

genes was implemented by the GOseq R package (Young et al.,

2010), in which gene length bias was corrected GO terms with a

corrected P value less than 0.05 were considered significantly

enriched KEGG (Kanehisa et al., 2008) (http://www.genome.jp/

kegg/) is a database resource for understanding high-level

func-tions and utilities of a biological system, such as a cell, an

organism or an ecosystem, from molecular-level information,

especially large-scale molecular datasets generated by genome

sequencing and other high-throughput experimental

technol-ogies We usedKOBASsoftware (Mao et al., 2005) to test the

statistical enrichment of differentially expressed genes in KEGG

pathways

Quantitative real-time PCR

To confirm the results of RNA-Seq analysis, the expression levels

of randomly selected genes were measured by quantitative

real-time (qRT)-PCR The primer sequences are summarized in

Table S1 The qRT reaction was performed using 5μg of total

RNA for each sample and a random 9-mer primer mix, used

according to the manufacturer’s instructions (Surcel Biotech,

China) The qRT-PCR reactions were performed on the Applied

Bio-systems 7300 Real-Time PCR System with the SYBR Premix

Ex Taq (Takara Bio, Kyoto, Japan) in eight connected tubes

(Takara) Each sample was analysed in triplicate in a 20-μl total

reaction volume, containing 5 pmol of each primer, 12.5μl SYBR

Green and 2μl diluted cDNA The relative expression levels were

calculated using the 2-ΔΔCtmethod (Livak & Schmittgen, 2001)

Data accessibility

The transcriptome data used in the present study are available in

the ArrayExpress database (http://www.ebi.ac.uk/arrayexpress)

under accession number E-MTAB-2491

Acknowledgements

We thank Ya-Ting Chen, Jia-Fei Ju and Peng-Yu Jin of the

Department of Entomology, Nanjing Agricultural

Univer-sity, for their help with the experiment This study was

supported in part by a grant-in-aid from the Science and Technology Programme of the National Public Welfare Professional Fund (201103020) from the Ministry of Agri-culture of China, and a grant-in-aid for Scientific Research (31172131 and 30871635) from the National Natural Science Foundation of China, and a grant-in-aid for Inno-vation Project (CXZZ13_0303) from Jiangsu Province, China.

References

Anders, S and Huber, W (2011) HTSeq: analysing high-throughput sequencing data with Python http://www-huber.embl.de/users/anders/HTSeq/ Accessed on 7 June 2013

Anders, S and Huber, W (2012) Differential Expression of

RNA-Seq Data at the Gene Level the Deseq Package EMBL,

Heidelberg

Benjamini, Y and Hochberg, Y (1995) Controlling the false dis-covery rate: a practical and powerful approach to multiple

testing J R Statist Soc B 57: 289–300.

Bian, G., Xu, Y., Lu, P., Xie, Y and Xi, Z (2010) The

endosymbiotic bacterium Wolbachia induces resistance to

dengue virus in Aedes aegypti Plos Pathog 6: e1000833.

Bolland, H.R., Gutierrez, J and Flechtmann, C.H.W (1998)

World Catalogue of the Spider Mite Family (Acari: Tetranychidae), with References to Taxonomy, Synonymy, Host Plants and Distribution Brill Academic press, Leiden.

Bourtzis, K., Pettigrew, M.M and O’Neill, S.L (2000) Wolbachia

neither induces nor suppresses transcripts encoding

antimi-crobial peptides Insect Mol Biol 9: 635–639.

Breeuwer, J.A.J (1997) Wolbachia and cytoplasmic incompatibil-ity in the spider mites Tetranychus urticae and T turkestani.

Heredity 79: 41–47.

Brennan, L.J., Keddie, B.A., Braig, H.R and Harris, H.L (2008)

The endosymbiont Wolbachia pipientis induces the expres-sion of host antioxidant proteins in an Aedes albopictus cell

line PLoS ONE 3: e2083.

Brennan, L.J., Haukedal, J.A., Earle, J.C., Keddie, B and Harris, H.L (2012) Disruption of redox homeostasis leads to oxidative DNA damage in spermatocytes of Wolbachia-infected

Drosophila simulans Insect Mol Biol 21: 510–520.

Bryon, A., Wybouw, N., Dermauw, W., Tirry, L and Van Leeuwen,

T (2013) Genome wide gene-expression analysis of faculta-tive reproducfaculta-tive diapause in the two-spotted spider mite

Tetranychus urticae BMC Genomics 14: 815.

Chevalier, F., Herbinière-Gaboreau, J., Charif, D., Mitta, G.,

Gavory, F., Wincker, P et al (2012) Feminizing Wolbachia:

a transcriptomics approach with insights on the immune

response genes in Armadillidium vulgare BMC Microbiol 12

(Suppl 1): S1

Chudzinski-Tavassi, A.M., Carrijo-Carvalho, L.C., Waismam, K., Farsky, S.H.P., Ramos, O.H.P and Reis, C.V (2010) A

lipocalin sequence signature modulates cell survival FEBS

Lett 584: 2896–2900.

Clark, M.E., Heath, B.D., Anderson, C.L and Karr, T.L (2006) Induced paternal effects mimic cytoplasmic incompatibility in

Drosophila Genetics 173: 727–734.

Darby, A.C., Armstrong, S.D., Bah, G.S., Kaur, G., Hughes, M.A.,

Kay, S.M et al (2012) Analysis of gene expression from the